Skip to main content

Full text of "USPTO Patents Application 09773351"

See other formats


REVIEW 

See related Commentary on page iv 

Barrier Function of the Skin: "La Raison d'Etre" of the Epidermis 



Kathi G Madison 

Marshall Dermatology Research Laboratories, Department of Dermatology, University of Iowa, Roy J. and Lucille A. Carver College of Medicine, 
Iowa City, Iowa, USA 



The primary function of the epidermis is to produce 
the protective, semi-permeable stratum corneum that 
permits terrestrial life. The barrier function of the stra- 
tum corneum is provided by patterned lipid lamellae 
localized to the extracellular spaces between corneo- 
cytes. Anucleate corneocytes contain keratin filaments 
bound to a peripheral comified envelope composed of 
cross-linked proteins. The many layers of these specia- 
lized cells in the stratum corneum provide a tough and 
resilient framework for the intercellular lipid lamellae. 
The lamellae are derived from disk-like lipid mem- 
branes extruded from lamellar granules into the inter- 
cellular spaces of the upper granular layer. Lysosomal 
and other enzymes present in the extracellular com- 
partment are responsible for the lipid remodeling re- 
quired to generate the barrier lamellae as well as for 
the reactions that result in desquamation. Lamellar 
granules likely originate from the Golgi apparatus and 
are currently thought to be elements of the tubulo-vesi- 
cular trans-Golgi network. The regulation of barrier li- 
pid synthesis has been studied in a variety of models, 
with induction of several enzymes demonstrated during 



fetal development and keratinocyte differentiation, but 
an understanding of this process at the molecular gene- 
tic level awaits further study. Certain genetic defects in 
lipid metabolism or in the protein components of the 
stratum corneum produce scaly or ichthyotic skin with 
abnormal barrier lipid structure and function. The in- 
flammatory skin diseases psoriasis and atopic dermatitis 
also show decreased barrier function, but the underly- 
ing mechanisms remain under investigation. Topically 
applied "moisturizers" work by acting as humectants 
or by providing an artificial barrier to trans-epidermal 
water loss; current work has focused on developing a 
more physiologic mix of lipids for topical application 
to skin. Recent studies in genetically engineered mice 
have suggested an unexpected role for tight junctions 
in epidermal barrier function and further developments 
in this area are expected. Ultimately, more sophisticated 
understanding of epidermal barrier function will lead 
to more rational therapy of a host of skin conditions in 
which the barrier is impaired. Key words: ceramides/ 
desquamation/ Golgi/keratinocyte/lipid. J Invest Dermatol 
121:231-241, 2003 



Life on dry land requires the presence of a barrier to 
water loss to prevent desiccation (Attenborough, 1980). 
That the skin provided this barrier was intuitively 
obvious, but it was not until the 1940s that the stratum 
corneum (SC) clearly emerged as the specific site 
of this barrier (Winsor and Burch, 1944; Blank, 1953). Although 
the typical "basket-weave" appearance of the SC in routine histo- 
logic sections does not give the impression that it could function 
as an effective barrier, this is an artifact of tissue processing. In 
fact, as can be seen on frozen sections of the epidermis and 
in fortuitous electron microscopic sections (Fig 1), the corneo- 
cytes are tightly opposed to each other. The barrier to water 
permeation is not absolute and the normal movement of water 
through the SC into the atmosphere is known as transepidermal 
water loss (TEWL) and constitutes part of insensible water 



Manuscript received May 29, 2002; revised February 12, 2003; accepted 
for publication March 16, 2003 

Address correspondence and reprint requests to: Kathi C Madison, MD, 
Department of Dermatology, University of Iowa Hospital, Iowa City, 
Iowa 52242, USA. Email: kathi-madison@uiowa.edu 

Abbreviations: LG, lamellar granule; SC, stratum corneum; TEWL, 
transepidermal water loss; AcylGlcCer, acylglucosylceramide(s); NMF, nat- 
ural moisturizing factor. 



loss. The SC is also the principal barrier to the percutaneous 
penetration of exogenous substances, both accidentally encoun- 
tered as well as deliberately applied. Epidermal barrier function 
and the related field of percutaneous absorption have been active 
areas of investigation in both academia and industry for many 
years; the information presented in this review is focused on the 
water barrier function of the epidermis and is intended as an 
overview, highlighting key points with relevance to both clini- 
cians and basic scientists. 



LIPIDS COMPRISE THE PERMEABILITY BARRIER 

In the 1950s and 1960s, experiments were done showing that sol- 
vent extraction of epidermis dramatically increased water perme- 
ability, implicating lipids in cutaneous barrier function (Berenson 
and Burch, 1951; Onken and Moyer, 1963; Matoltsy et al, 1968; 
Scheuplein and Ross, 1970; Sweeny and Downing, 1970). Although 
some earlier studies had noted the pronounced changes in lipid 
composition that accompany keratinocyte differentiation, it was 
not until the pioneering studies of Gray, Yardley, and colleagues 
in the 1970s that an accurate picture of epidermal and SC lipid 
composition was established (reviewed in Yardley and Summerly, 
1981). Thin layer chromatographic analysis of the solvent extracta- 



0022-202X/03/$15.00 - Copyright © 2003 by The Society for Investigative Dermatology, Inc. 

231 



232 MADISON 



THE JOURNAL OF INVESTIGATIVE DERMATOLOGY 




Ceramide 1 [EOS] 




Figure 1. Electron micrograph showing the upper stratum granulo- 
sum and SC of an organotypic mouse keratinocyte culture. This fig- 
ure shows a portion of a micrograph previously published in Madison et al 
(1988). Arrows, keratohyalin granules; arrowheads, intercellular spaces be- 
tween closely opposed comeocytes; stars, artifactual separation. The inter- 
cellular spaces appear empty in this osmium tetroxide postfixed specimen. 
Original magnification X 6000. 



ble lipids from SC reveals an unusual lipid composition consisting 
of a roughly equimolar mixture of ceramides (45-50% by weight), 
cholesterol (25%), and free fatty acids (10-15%) plus less than 5% 
each of several other lipids, the most important of which is choles- 
terol sulfate. The detailed structures of the ceramide species of pig, 
mouse, and human skin were determined in the 1980s (Wertz and 
Downing, 1983; Long et al, 1985; Madison et al, 1990) with refine- 
ments still being published (Robson et al, 1994; Doering et al, 
1999a; Stewart and Downing, 1999; Hamanaka et al, 2002; 
Ponec et al, 2003). Human SC ceramide structures are shown in 
Fig 2. 




Ceramide? [AP] 

0 



Ceramide 8 [AH] 




Ceramide 9 [EOP] 



Figure 2. Structures of the free ceramides of human SC. Numbers 1 
to 8 represent thin layer chromatographic mobility with ceramide 1 being 
the least polar and ceramide 8 the most polar. Ceramide 9 (EOP) has re- 
cently been discovered (Ponec et al, 2003) and has a thin layer chromato- 
graphic mobility between that of ceramide 2 and ceramide 3. The letters 
in parentheses give the ceramide classification as suggested in Motta et al 
(1993). 

STRATUM CORNEUM LIPIDS ARE ORGANIZED AS . 
STACKED MEMBRANE SHEETS IN THE 
INTERCELLULAR SPACES 

Early freeze fracture electron micrographic studies of the epider- 
mis had demonstrated the presence of broad continuous lipid 
sheets in the extracellular spaces of the SC (Breathnach et al, 1973; 
Ehas and Friend, 1975; Elias et al, 1977), but these lipid membranes 
were not visible in conventional electron microscopy. Fixation 
with ruthenium tetroxide, however, which is more reactive 
than the usual osmium tetroxide fixative, clearly demonstrates 
the stacked and patterned lipid sheets in the extracellular spaces 
of the SC (Madison et al, 1987; Fig 3). All of the free fatty acids 
and the amide-linked fatty acid chains in the ceramides are non- 
branched and have no double bonds. This allows for tight lateral 
packing and the formation of highly ordered gel phase membrane 
domains, which are less fluid and less permeable than typical li- 
quid crystalline phospholipid-dominant biologic membranes. 
Cholesterol may provide some necessary fluidity to the mem- 
branes, which might otherwise be too rigid and possibly brittle. 
Numerous biophysical studies of SC structure suggest the pre- 
sence of coexisting liquid crystalline and gel phase domains in 
the membranes of the SC. This concept was suggested by Fonlind 
(1994) and presented as the "domain mosaic" model; recently a new 
model for the existence of fluid phases within the lamellae, the 
"sandwich model" was presented by Bouwstra et al (2000). Norlen 
(2001b), however, has very recently proposed a different "single gel 
phase" model that he feels is more consistent with the documented 
barrier properties of the SC. There are still many unanswered 
questions about the exact way in which the SC lipids are orga- 
nized at the molecular level and this is an active area of research 
(reviewed in Bouwstra et al, 2003), Understanding the physical 
structure of the membranes is critical to understanding their func- 
tion as a barrier, both to water and to other substances, and ulti- 
mately to understanding the mechanisms of barrier disruption in 
a variety of skin diseases. 



VOL. 121, NO. 2 AUGUST 2003 



EPIDERMAL BARRIER FUNCTION 233 



STRATUM CORNEUM BARRIER LIPIDS ORIGINATE 
FROM LAMELLAR GRANULES 

Lamellar granules (LG) are small organelles with a bounding 
membrane, most prominent in the granular cell layer of the epi- 
dermis and visible only by electron microscopy. They contain 
stacks of lipid lamellae (Fig 4a) composed of phospholipids, cho- 
lesterol, and glucosylceramides (Freinkel and Traczyk, 1985) that 
are the precursors of the SC intercellular lipids (reviewed in 
Landmann, 1988). Late in epidermal differentiation, at the transi- 
tion from granular cell to comeocyte, LG are thought to fuse 
with the plasma membrane of the granular cell and discharge 
their lipid membranes into the intercellular space (Fig 4b). Along 
with the lipids, LG secrete a group of acid hydrolases (Freinkel 
and Traczyk, 1985; Grayson et al, 1985; Menon et al, 1986. 1992), 
which break down the phospholipids and convert glucosylcera- 
mides to ceramides. The enzyme responsible for the latter reac- 
tion (Holleran et al, 1994), p-glucocerebrosidase, is the enzyme 
defective in Gaucher disease. Although in most Gaucher patients 
residual enzyme activity is sufficient to catalyze the cutaneous re- 
action, there is a subset of patients with severe enzyme deficiency 
who present as collodion babies, have abnormal barrier function, 
and die in the neonatal period (Sidransky et at, 1992). 

Other enzymes involved in the lipid metabolic changes that 
occur after extrusion of LG contents include acid sphingomyeli- 
nase and secretory phospholipase A2, both of which have been 
shown to be required for permeability barrier function (Jensen 
et al 1999; Elias et a/, 2000; Schmuth et al, 2000). Lysosomal acid, 
lipase activity is also present in LG (Madison et d, 1998), but its 
exact function has not been determined. Some of the proteases 
that appear to regulate desmosome breakdown and contribute to 
desquamation (see below) may also be delivered via LG (Sondell 
et al 1995). 

Concurrent with the complex changes in lipid composition 
that occur following extrusion of LG contents, the short stacks 
of membranes reorganize structurally to form patterned lamellar 
sheets as shown in Fig 3. This transformation has been suggested 
to occur via edge-to-edge fusion of the lipid stacks (Landmann, 
1986) and calcium may promote this process (Abraham et al, 1987). 
Based on knowledge of the lipid composition of the lamellae and 
their electron microscopic appearance in ruthenium tetroxide- 
fixed sections, a model for their biochemical structure has been 
proposed (Swartzendruber et al, 1989). 

LIPIDS UNIQUE TO KERATINIZING EPITHELIA MAY 
PLAY A SPECIFIC ROLE IN MEMBRANE FORMATION 

LG are particularly enriched in a lipid unique to keratinizing 
epithelia, acylglucosylceramide (AcylGlcCer), This unusual lipid 




Figures. Electron micrograph showing the stacked and patterned 
lamellar membrane sheets in a single intercellular space in mouse 
SC postfixed with ruthenium tetroxide. The cornified envelope of the 
lower corneocyte is clearly visible {arwws)\ ICS, intercellular space; K, ker- 
atin contents of the comeocytes bordering the intercellular space. Original 
magnification X 200,000, 




Figure 4, (a) Left: electron micrograph of a single LG in mouse 
epidermis. A lower magnification view of the same LG was used in a pre- 
viously published figure (Madison et al, 1987). Arrows, bounding mem- 
brane. Original magnification X 300,000. Right: a schematic diagram of a 
LG as suggested by Landmann (1986). (£») Electron micrograph of extruded 
LG contents in the intercellular space at the junction of the granular layer 
and the SC. G, granular cell; ICS, intercellular space; Arrows, granular cell 
plasma membrane; open arrow, LG. Original magnification X 125,000. 



has a very long chain CO-hydroxy fatty acid moiety (C28-36) with 
linoleic acid (an essential fatty acid) ester-linked to the oo-hydro- 
xyl group (Abraham et al, 1985; Fig 5, top). The interior lipid la- 
mellae of LG have been suggested to arise from the flattening and 
stacking of lipid vesicles (Landmann, 1986) and AcylGlcCer has 
been proposed to function as a molecular rivet to accomplish this 
process (Wertz and Downing, 1982; Fig 5, bottom). The fatty acid 
chain is long enough to span completely a Hpid bilayer and allow 
the linoleate tail to insert into a leaflet of a neighboring bilayer as 
shown in Fig 5 (bottom). Figure 4(a, right) schematically shows 
the interior structure of a LG as flattened and stacked lipid vesi- 
cles. Evidence to support this model includes the ability of 
AcylGlcCer to cause the flattening and aggregation of lipid lipo- 
somes in vitro (Landmann et al, 1984). Menon et al (1992) have 
suggested an alternative model of accordion-like pleating of lipid 
membranes to explain the appearance of LG contents followed by 
"unfurUng" after extrusion. AcylGlcCer could function as a rivet 
in this model as well. As the biophysics of membrane dynamics 
and the function of the Golgi apparatus (see below) are better un- 
derstood, new models of LG assembly/extrusion may well 
emerge. After extrusion of LG contents, AcylGlcCer is deglyco- 
sylated, along with the rest of the glucosylceramides, to produce 
acylceramide (Fig 6, middle structure). Acylceramide is thought 
to perform the same molecular rivet role in the SC lamellae as 
AcylGlcCer does in the LG, and there are X-ray diffraction data 
to support this concept (Schreiner et al, 2000). 

Acylceramide and its precursor are the two principal carriers of 
linoleic acid in the SC and living epidermis, respectively It has 
been known for years that essential fatty acid deficiency results 
in poor cutaneous barrier function and increased TEWL (re- 
viewed in Wertz et al, 1987). These effects correlate with replace- 
ment of linoleic acid by oleic acid in AcylGlcCer and 
acylceramide (Melton et al, 1987), a substitution that results in al- 



234 MADISON 



THE JOURNAL OF INVESTIGATIVE DERMATOLOGY 




AcylGlcCer 




Lipid yesklewith 



iKtiiHciied vesicle s dmhit lrilii$%rdisk 



Figure 5. Top: structure of acylglucosylceramide. Bottom: proposed 
model of acylglucosylceramide function in the flattening and stacking of 
lipid vesicles to generate the double bilayer structure of the internal LG 
lamellae. The orientation of acylglucosylceramide is not known and both 
possibihties are shown. The vesicle and double bilayer disk are shown in 
cross-section. Arrowheads, linoleate moiety; Glu, glucose. 

tered biophysical properties of the SC lamellae (Bouwstra et al, 
2002) and increased water permeability. 



THE LIPID ENVELOPE 

Each comeocy te has an approximately 10 nm thick tough periph- 
eral protein envelope, called the comified envelope, that is com- 
posed of several structural proteins, notably involucrin and 
loricrin, cross-linked by sulfhydryl oxidases and transglutami- 
nases (reviewed in Kalinin et at, 2002). The interior surface of the 
comified envelope is linked to the bundles of keratin filaments 
that fill the intracellular compartment of comeocy tes. The multi- 
ple layers of comeocy tes in the SC contribute a tough and resili- 
ent framework for the intercellular lipid lamellae. On the exterior 
(extracellular) surface of the comified envelope is a covalently 
bound layer of very long chain (O-hydroxyceramides called the 
lipid envelope (Swartzendruber et al, 1987; Wertz and Downing, 
1987; Fig 6, bottom). This structure can be seen on electron mi- 
croscopy of SC that has been solvent extracted to remove all of 
the free lipid, as shown in Fig 7. Evidence suggests that the co- 
hydroxyceramides are ester-linked to involucrin amino acid resi- 
dues (Downing, 1992; Marekov and Steinert, 1998) and transglu- 
taminase has been shown to be capable of catalyzing this reaction 
(Nemes et al, 1999). Specific three-dimensional conformations for 
involucrin that would allow for attachment of lipids on the exter- 
ior surface and other envelope proteins on the interior surface of 
the comified envelope have been proposed (Lazo and Downing, 
1999; Kajava, 2000). Surprisingly, however, both involucrin 




a)-Hydroxyceramid9 



Figure 6. The very long chain CD-hydroxy fatty acid-containing cer- 
amides of mammalian epidermis. LG acylglucosylceramides are the 
precursors of the acylceramides (ceramide 1 (EOS) is shown) in the SC in- 
tercellular lamellae and the oa-hydroxyceramides of the lipid envelope. See 
text for details. 



Figure % Electron micrograph of mouse SC that has been solvent 
extracted to remove all of the free lipid. A lucent band (the lipid en- 
velope) remains on the exterior surface of the comified envelopes of adja- 
cent comeocytes. In solvem-extraaed SC the lipid envelopes are tightly 
opposed and a narrow dark line can be seen where they join (see Wertz et 
alt 1989 for a more detailed discussion). The dominant component of the 
lucent band is very long chain CO-hydroxyceramides derived from LG acyl- 
glucosylceramides (likely from the LG bounding membrane; see text and 
Fig 6) and covalently bound to comified envelope proteins. K, keratin con- 
tents of two adjacent comeocytes; arrowheads, comified envelopes. Ruthe- 
nium tetroxide postfixation, original magnification X 125,000. 



(Djian et at, 2000) and loricrin (Koch et at, 2000; Jamik et ah 2002) 
knockout mice have relatively normal-appearing comified envel- 
opes and no epidermal phenotype, suggesting great redundancy 
in the components of the epidermal barrier (Steinert, 2000). This 
also implies that other envelope proteins are able to bond with a>- 
hydroxyceramides and, indeed, new envelope-associated proteins 
continue to be discovered (Cabral et al, 2001; Marshall et al, 2001). 

AcylGlcCer, the precursor of acylceramide (see above), is also 
the precursor of the (O-hydroxyceramides of the lipid envelope. 
Wertz (1996) has found that two-thirds of LG AcylGlcCer is in 
the bounding membrane; this suggests that lipid envelope to-hy- 
droxyceramides are delivered to the cell surface when LG bound- 
ing membranes fuse with the granular cell plasma membrane 
(Wertz, 1996; Kalinin et al, 2002). Recent evidence suggests that 
most or all or the (D-hydroxyceramides are bound by their a>-hy- 
droxyl ends (Nemes et al 1999; Doering et al, 1999b; Stewart and 
Downing, 2001), This implies that the linoleic add tail must be 
removed from the CG-hydroxyl end of AcylGlcCer; the enzyme 
responsible for this deacylation is not known, but candidates in- 
clude transglutaminase (Nemes et al, 1999; Kalinin et al, 2002) and 
acid lipase. The uniquely long fatty add chains of the lipid en- 
velope ceramides span the distance of a typical plasma membrane 
bilayer leaving the sphingosine chains free to interdigitate with 
the nonbound intercellular lipid lamellae. This chain interdigita- 
tion may contribute to the patterned organization of the lamellae 
seen on electron microscopy. The structures of the very long 
chain Ci>-hydroxy fatty acid-containing ceramides are shown in 
Fig 6. Note that human ceramide 4 and the newly discovered cer- 
amide 9 (Fig 2) also contain a very long chain ©-hydroxy fatty 
acid and contribute co-hydroxyceramides to the lipid envelope. 
Differentiation-related changes in lipid structures are illustrated 
schematically in Fig 8. 



DESQUAMATION 

It has long been known from the clinical example of X-linked 
ichthyosis, caused by cholesterol sulfatase defidency (Shapiro 
et al, 1978), that the hydrolysis of cholcstgrol sulfate in the SC is 
important for comeocyte desquamation . The mechanism by 
which excess cholesterol sulfate inhibits desquamation and its hy- 
drolysis promotes desquamation , however, is still under investi- 
gationr Excess cholesterol sulfate has been suggested to alter the 
structure and function of the lipid bilayers (Zettersten et al, 1998; 



^Sando GN, Howard EJ, Madison KC: Add lipase expression in cultured 
human keratinocytes: Potential role in epidermal ceramide metabolism. 
J Invest Dermatol 108:554a, 1997 (Abstr.) 



VOL. 121, NO. 2 AUGUST 2003 



EPIDERMAL BARRIER FUNCTION 235 




Figure 8. Schematic diagram of the epidermis, including the trans- 
formations of lipid structures that accompany epidermal differen- 
tiation. Keratin filaments are omitted. Not to scale. 

Bouwstra et al, 1999). Proteases, especially SC tryptic enzyme and 
SC chymotryptic enzyme, have been implicated m desmosome 
breakdown and comeocyte desquamation (Hansson et al, 1994; 
Honkoshi et al, 1998; brattsand and Egelrud, 1999; Simon et al, 
2001), and cholesterol sulfate has been shown to inhibit some of 
their activities (Sato et al, 1998). A recent study showed that cho- 
lesterol sulfate inhibits transglutaminase-mediated involucrin 
cross-linking as well as involucrin esterification to the co-hydro- 
xyceramides of the lipid envelope (Nemes et at, 2000). Any of 
these mechanisms could contribute to the so-called "retention hy- 
perkeratosis" of X-linked ichthyosis. 



REGULATION OF EPIDERMAL LIPID SYNTHESIS 

Epidermal barrier lipid synthesis has been studied in vivo in adult 
pigs (Hedberg et al 1988; Wertz and Downing, 1990) and hairless 
mice (reviewed in Feingold and Elias, 2000), in fetal rat develop- 
ment models in vivo and in vitro (reviewed in Williams et al, 1998), 
during fetal mouse development (Doering et al, 2002), and in a 
variety of keratinocyte culture models (Madison et al, 1989, 1990; 
Jetten et at, 1992; Sando et al 1996; Ponec et al, 1997; Watanabe et at, 
1998). We know from in vivo metabolic labeling studies that bar- 
rier lipids are largely synthesized de novo from acetate (Hedberg 
et al 1988; Wertz and Downing, 1990). The loss of phospholipids 
and the conversion of glucosylceramides to ceramides and 
AcylGlcCer to acylceramide and (0-hydroxyceramides during 
epidermal differentiation have been demonstrated in vivo as well 
as in vitro (Hedberg et al 1988; Madison et al, 1990). As an essential 
fatty acid, linoleate must be derived from the circulation, but 
may also be recycled within the epidermis (Madison et al 1989; 
Wertz and Downing, 1990). 

Several of the enzymes known to be required for either barrier 
lipid synthesis or postextrusion processing are induced during 
keratinocyte differentiation in vitro (Jetten et al 1992; Sando et al 
1996; Watanabe et al 1998) or during fetal rat epidermal develop- 
ment (Williams et al 1998). Submerged explant cultures of fetal 
rat skin recapitulate in utero barrier development and the develop- 
ment of the barrier can be inhibited by testosterone and stimu- 
lated by exogenous application of glucocorticoids, thyroid 
hormone, and estrogen and by lifting the cultures to the air-li- 
quid interface (Williams et al 1998). More recently, peroxisome 
proliferator activated receptor (PPARa) ligands (clofibrate, lino- 
leic add, oleic acid) and farnesol, a metabolite in the cholesterol 
biosynthetic pathway, have been shown to stimulate epidermal 
differentiation and barrier development in fetal rats both in utero 
and in skin explant cultures (Hanley et al, 1999). PPAR are nuclear 
hormone receptors that heterodimerize with retinoid X receptors 
to regulate the transcription of several genes involved in lipid me- 



tabolism and have been most studied in adipocytes. Recently, 
mice transgenic for a dominant-negative mutant retinoic acid 
receptor a were shown to have markedly decreased barrier func- 
tion associated with abnormal lipid processing, implicating reti- 
noid receptor-mediated signaUng pathways in barrier formation 
(Attar et al 1997). Which genes involved in epidermal barrier lipid 
synthesis or processing might be regulated by these hormones 
and receptors is not known, although increased p-glucocerebro- 
sidase and cholesterol sulfatase activity following treatment with 
PPAR ligands has been shown (Hanley et al 1999). PPAR-a 
knockout mice show a delay in fetal SC formation as well as 
decreased [3-glucocerebrosidase activity, but are normal by the 
time of birth (Schmuth et al 2002). Another study in transgenic 
mice has demonstrated that a member of the kruppel family of 
transcription factors, Klf4, is essential for normal barrier develop- 
ment (Segre et al 1999). These and other mouse models should be 
of great help in dissecting the mechanisms of barrier formation at 
the molecular genetic level. 

Artificial barrier disruption of hairless mouse skin by acetone 
wiping or tape stripping (as measured by increased TEWL) has 
been used as a model to study the events involved in barrier 
repair. The most commonly used approach to determine if an 
effect is mediated by loss of barrier rather than a nonspecific 
injury effect is to immediately cover one site with a vapor 
impermeable membrane and compare the response to an uncov- 
ered site. Studies using this model have shown acute barrier dis- 
ruption to stimulate epidermal proliferation and increase mRNA 
levels and activities of several (but not all) of the enzymes in- 
volved in barrier lipid synthesis, in particular those associated 
with fatty acid, cholesterol, and ceramide synthesis (Feingold 
and Elias, 2000). In general, vapor-impermeable and semiperme- 
able membranes inhibit the stimulatory response and delay bar- 
rier repair, the degree of inhibition correlating with the degree 
of impermeability The molecular mechanisms by which barrier 
disruption produces, and artificial barrier restoration inhibits, 
these effects remain unknown, although changes in SC water 
content (Denda et al 1998; Fluhr et al 1999), ion content and dis- 
tribution (particularly calcium) (Lee et al 1998), or cytokine pro- 
duction (Jensen et al 1999) may be involved in the signaling 
pathways. 

Although a recent study showed an increase in serine 
palmitoyltransferase (the rate-limiting enzyme in sphingolipid 
synthesis) mRNA following tape stripping of human skin 
(Stachowitz et al, 2002), as rodent epidermis is different from 
human, including having poorer barrier function, which of 
the findings in hairless mouse models will translate to human 
skin needs further study (Rigg and Barry, 1990, and references 
therein). Several studies of human skin in vivo have shown no ef- 
fect of occlusion with membranes of varying permeability 
on barrier recovery following tape stripping, detergent-induced 
damage, or wounding (Silverman et al 1989; Van de Kerkhof 
et al 1995; Welzel et al 1995, 1996; Fluhr et al 1999) and a study in 
premature infants showed that the use of semipermeable mem- 
branes improved barrier function in treated compared with un- 
treated sites (Mandni et al 1994). Under physiologic conditions, 
barrier lipid synthesis, LG formation, and lipid extrusion take 
place continuously under a competent barrier. Whether barrier 
abrogation results in additional stimulation of all of these pro- 
cesses, as has been demonstrated in mouse skin, needs additional 
study in human skin. 



LAMELLAR GRANULE ASSEMBLY 

An unanswered question in the field of epidermal barrier 
formation is how keratinocyte LG assembly is orchestrated. Con- 
taining both lipid membranes and acid hydrolases destined for 
extrusion into the extracellular environment, the LG is some- 
thing of a cross between a secretory granule and a lysosome. 
A large body of evidence now supports the concept that LG 



236 MADISON 



THE JOURNAL OF INVESTIGATIVE DERMATOLOGY 



originate from the Golgi apparatus and the very active and ra- 
pidly advancing field of Golgi research is ripe for application 
to keratin ocyte biology Of particular interest is recent work on 
the trans-Golgi network, which is the highly tubulated sorting 
and delivery portion of the Golgi apparatus. It is now thought 
that Golgi to plasma membrane transport is mediated by 
pleiomorphic tubulovesicular structures (sometimes referred to 
as "post-Golgi carriers") that are formed by maturation of the 
trans-Golgi compartment, rather than by vesicles (Hirschberg 
et al, 1998; Mironov et at, 1998). In this paradigm, secretory orga- 
nelles are the remnants of Golgi cistemae that have already ex- 
ported all of the components not destined for secretion; thus 
they are formed by terminal maturation of trans-Golgi network 
cisternae. 

Careful examination of high magnification electron 
microscopy images of epidermis clearly shows that LG do not 
constitute a uniform vesicular population. There are numerous 
highly irregular shapes, including ovals, dumbbells, and elon- 
gated tubular structures filled with the characteristic stacked 
lamellae. These images are consistent with sections through a 
tubular network (Madison and Howard, 1996; Elias et at, 1998; 
Madison et al, 1998) suggesting that keratinocyte lamellar "gran- 
ules" are trans-Golgi network structures and that the secretion of 
their contents may be an excellent example of the current Golgi 
paradigm. Further studies are needed to determine the validity of 
this paradigm and/or whether the specific mechanisms may be 
unique to keratinizing epithelia. Norlen (2001a) has recently pro- 
posed a "membrane folding model" of barrier lipid delivery that 
does not require membrane trafficking or fusion as classically 
described. 

Even if we accept the Golgi origin of lamellar "granules", how 
all of the enzymes involved in barrier lipid synthesis are regu- 
lated, how the internal membranes are formed, how lysosomal 
enzymes are incorporated into the membrane structure, what sti- 
mulates fusion (if classic membrane fusion does occur) with the 
keratinocyte plasma membrane, and how the whole process is co- 
ordinated with the many other events occurring during terminal 
epidermal differentiation are questions that remain to be an- 
swered. Clearly, disruptions in any of these processes could have 
a significant effect on SC barrier function. 



FLAKY SKIN 

Flaky skin, often called "dry" skin, is a cutaneous reaction pattern 
reflecting abnormal desquamation of diverse etiologies. Comeo- 
cytes are normally shed in small enough groups that they are not 
visible on the skin surface; when this process is disturbed in any 
way, comeocytes collect in visible clumps (scales) that produce a 
rough texture and appearance. 

The importance of SC water content to "normal" nonflaky skin 
appearance has long been known, with healthy tissue containing 
greater than 10% water (Blank, 1952, 1953). Both water soluble in- 
tracomeocyte substances (collectively referred to as natural moist- 
urizing factor (NMF)) (reviewed in Harding et al, 2000) and the 
intercellular lipid membranes (Imokawa et al, 1991) contribute to 
the water binding properties of the SC and the barrier properties 
of the intercellular membranes maintain hydration by limiting 
water loss from the tissue. NMF consists of a mixture of amino 
acids and their derivatives (pyrrolidone carboxylic acid, urocanic 
acid), lactic acid, urea, and sugars that is highly hygroscopic and 
acts as an endogenous humectant. The amino acid portion of 
NMF derives from proteolysis of filaggrin (from keratohyalin 
granules) in the mid to outer SC (Harding et at, 2000). One of 
the critical functions of water in the SC is participating in the 
many hydrolytic enzymatic processes required for normal des- 
quamation (discussed above) and for the generation of NMF. 

Knowing this, we can predict that any endogenous defect (pri- 
mary or secondary) or exogenous insult that decreases SC NMF 
content, alters the composition or physical properties of the inter- 



cellular lipids, or disrupts epidermal differentiation may lead to 
improper desquamation and clinical scaling. This is a simplifica- 
tion of a very complex situation, but helps in thinking about 
some mechanisms that can lead to "flaky" skin. Examples include 
ichthyosis vulgaris, where there is a profound deficiency in filag- 
grin (Sybert et a/, 1985) (and thus NMF), aged skin where lipid 
syndiesis (particularly cholesterol) is decreased leading to poor 
barrier repair after insults (Ghadially et at, 1995, 1996a), and so- 
called "winter xerosis" where low environmental humidity de- 
creases SC water content. Whether "flaky skin" will have impaired 
barrier function depends on the underlying pathophysiology as 
well as the effect of any compensatory response. Patients with la- 
mellar ichthyosis due to transglutaminase 1 deficiency have ab- 
normal comified envelope structure and dramatically scaly skin. 
Their impaired barrier function has been reported to be due to 
defects in the composition and organization of the SC lipid la- 
mellae (Lavrijsen et al, 1995; Pilgram et at, 2001; Elias et al, 2002) 
directly related to the underlying disturbance in the comified en- 
velope (Elias et al, 2002). In epidermolytic hyperkeratosis, which 
is caused by genetic defects in the suprabasal keratins 1 and 10, 
keratinocytes are fragile and patients have both blistering and se- 
vere scaling. The barrier defect, however, has been shown to be 
due to abnormal LG secretion and the resulting decrease in SC 
lipid lamellae (Schmuth et al, 2001). In many inflammatory skin 
diseases where overall epidermal differentiation is disturbed, there 
are likely secondary effects on corneocyte structure and NMF 
generation as well as on the composition and function of the in- 
tercellular lipids that ultimately result in scaling. It should be 
noted here that sebaceous gland secretions (sebum) are unlikely 
to play a significant part in epidermal moisturizadon in humans; 
prepubertal children, who have essentially no sebum production, 
have enviable skin qualities and certainly no particular difficulties 
with xerosis. 



SKIN DISEASES WITH ABNOIUVIALITIES IN BARRIER 
FUNCTION AND IN LIPID METABOLISM 

There are several genetic skin diseases with known defects in li- 
pid meubolism that have scaly or ichthyotic skin as part of the 
clinical picture (reviewed in Williams and Elias, 2000). RXLI, 
discussed above, was the first of these diseases to be described. 
Others include Sjogren— Larsson syndrome (defect in fatty alde- 
hyde dehydrogenase; De Laurenzi et al, 1996), Refsum's disease 
(defect in phytanoyl-coenzyme A hydroxylase; Jansen et at, 1997), 
and X-linked dominant Conradi-Hunermann as well as CHILD 
syndrome (Congenital Hemidysplasia with /chthyosiform ery- 
throderma and Limb Defects; defects in 3p-hydroxysteroid-A 8, 
A7-isomerase, an enzyme in the sterol biosynthetic pathway; 
Braverman et at, 1999; Grange et al, 2000). CHILD syndrome can 
also be caused by mutations in the gene encoding 3p-hydroxys- 
teroid dehydrogenase (Konig et al, 2000). Very recently, Chanar- 
in-Dorfman syndrome (neutral hpid storage disease with 
ichthyosis), which is characterized by nonbullous congenital 
ichthyosiform erythroderma, was found to be caused by muta- 
tions in a gene (CGI-58) that encodes a new protein of unknown 
function in the esterase/lipase/thioesterase family (Lefevre et al, 
2001). Mutations in lipoxygenase-3 (ALOXE3) and 12R-lipoxy- 
genase (ALOX12B) have now been reported in nonbullous con- 
genital ichthyosiform erythroderma linked to chromosome 
17pl3.1 (Jobard et al, 2002). Although the substrates and products 
of these lipoxygenases are not yet known, further investigation 
should shed considerable light on the functional role of lipoxy- 
genase pathways in epidermal differentiation. For the majority of 
these diseases, even when the defect is known, the precise me- 
chanism by which SC structure and function are altered has not 
been determined. Harlequin ichthyosis is characterized by an ab- 
sence of LG and SC lipid lamellae (Milner et al, 1992). Some 
subtypes of congenital ichthyosiform erythroderma show 
abnormalities in lipid structures (Arnold et at, 1988) and diere are 



VOL 121, NO. 2 AUGUST 2003 



EPIDERMAL BARRIER FUNCTION 237 



some cases that share the microscopic findings reported for harle- 
quin ichthyosis (Virolainen et al, 2001), but the underlying defects 
in these genetic diseases remain unknown. 

Netherton syndrome and Papillon— Lefevre syndrome are 
caused by defects in proteolysis: mutations in SPINK5, a serine 
protease inhibitor (Chavanas et a/, 2000), and cathepsin C (Hart 
et al, 1999), respectively. Abnormalities in SC lipid structure have 
been described in Netherton syndrome (Fartasch et al, 1999), but 
hov^^ SP1NK5 defects result in these alterations is not knov^^n. A 
recent paper reported that SC hydrolytic activity is increased in 
Netherton syndrome and the authors suggest that SP1NK5 is ne- 
cessary to regulate the enzymes involved in desquamation (Ko- 
matsu et al, 2002). Type 2 Gaucher disease, with severe defects in 
p-glucocerebrosidase, discussed above, results in failure to meta- 
bolize LG glucosylceramides to ceramides. The recent finding of 
delayed barrier repair in Niemann-Pick disease (acid sphingo- 
myelinase deficiency) (Schmuth et aly 2000) suggests these patients 
might be more susceptible to exogenous barrier insults. Many ge- 
netic skin diseases with defects in a variety of epidermal protein 
structures, as discussed above for lamellar ichthyosis and epider- 
molytic hyperkeratosis, have associated changes in SC lipid struc- 
ture and function and these are currendy being investigated. 

Of the inflammatory skin diseases, atopic dermatitis and psor- 
iasis have been the most studied with respect to epidermal barrier 
function and SC lipid alterations. A decrease in ceramides has 
been the most consistent finding in atopic dermatitis and this 
has been suggested to result from increased sphingomyelin dea- 
cylase activity (Hara et al 2000). Macheleidt et al (2002) have re- 
cendy demonstrated decreased free very long chain fatty acids 
and lipid envelope (O-hydroxyceramides in atopic skin as well as 
decreased very long chain fatty acid, ceramide, and glucosylcera- 
mide synthesis by atopic epidermis in vitro. Although the mechan- 
isms underlying these changes are not known, the findings offer 
an explanation for the decreased barrier function of atopic skin. 
In addition, gene polymorphisms in SP1NK5, the gene mutated 
in Netherton syndrome (which has atopy as part of the clinical 
picture), have been found to show significant association widi 
common atopic disease, including atopic dermatitis (Walley et aly 
2001). Combined with the recent demonstration that mice over- 
expressing SC chymotryptic enzyme develop a chronic itchy der- 
matitis (Hansson et al, 2002), this suggests tantalizing links 
between excess protease activity, barrier function, epidermal dif- 
ferentiation, and inflammation. In psoriasis, alterations in cera- 
mide content have been demonstrated (Motta et al, 1994) and 
abnormal lipid structures reported (Ghadially et al, 1996b). More 
work is needed to determine whether these changes are primary 
or secondary and/or are specific for the disease. 



NEONATAL SKIN 

Current dogma holds that the SC is not functionally mature until 
32 to 34 wk estimated gestational age and that the skin of prema- 
ture infants develops competent barrier function within 2 to 4 wk 
of birth regardless of gestational age. A recent longitudinal study 
in premature infants suggests that the barrier may be mature at 
about 30 wk, but that for neonates younger than 25 wk estimated 
gestational age, postnatal barrier maturation may take as long as 5 
to 7 wk (Kali et a/, 1998). The high TEWL from premature infant 
skin can lead to multiple complications and a mainstay of therapy 
is to prevent this loss by maintaining a humidified environment 
or using petrolatum-based topical preparations until a competent 
barrier develops (Siegfried, 1998). In the future, this problem may 
be approached by stimulating the normal development of barrier 
function (either in uterOy if possible, or postnatally) and/or provid- 
ing a more physiologic temporary artificial barrier. Studies of fe- 
tal rat barrier development suggest that acceleration of this 
process is possible (Williams et al, 1998; Hanley et aly 1999), but this 
has not yet been shown in humans. In fact, a recent study of bar- 
rier maturation in preterm infants showed no stimulation by the 



administration of antenatal steroids and no difference between 
the sexes (Jain et aly 2000), although steroids stimulate, and testos- 
terone inhibits, barrier development in the rat model. If topically 
applied physiologic lipids must enter LG and be extruded to im- 
prove barrier function (see below), this approach may not work 
in premature skin where LG assembly and secretion might be 
relatively undeveloped. If a mix of appropriate lipids could be 
prepared such that they would function as they do in vivOy how- 
ever, repeated topical application should provide a physiologic 
barrier until normal epidermal maturation occurs. Any improve- 
ment in the current management of these fragile infants would be 
a valuable clinical contribution from basic work on epidermal 
barrier development and function. 



"MOISTURIZERS" AND BARRIER CREAMS 

The use of topically applied materials to improve the appearance, 
function, and "feel" of skin is as old as human life. Our knowl- 
edge about the mechanisms by which all of these agents work 
continues to evolve along with our understanding of SC struc- 
ture and function and at present is incomplete. As discussed 
above, flaky "dry" skin is a heterogeneous condition with varied 
pathogenetic mechanisms that may or may not be associated with 
altered barrier function. It follows that controlled studies in 
homogeneous populations of patients would be necessary to es- 
tablish the efficacy of an agent or product for a given condition. 

The term moisturizer implies that the substance applied adds 
water and/or retains water in the SC. This is true for many of 
the products in use today, although the mechanism by which this 
is accomplished may vary. In addition, "moisturizing" substances 
are known to have a variety of less well defined effects on SC 
function separate from their effects on water content. Urea, pro- 
pylene glycol, glycerin, and hydroxy acids (especially lactic add) 
are humectants (water holding) and are used in many moisturiz- 
ing formulations; however, they all also function as exfoliants, 
i.e., they promote desquamation. Although some of these sub- 
stances are referred to as "keratolytics" true protein denaturation 
only occurs at high concentrations that are not used in moisturiz- 
ing formulations. Whether the effect of these agents on desqua- 
mation is due solely to the increased water content or whether 
they have other effects on the desquamation process has not been 
completely worked out. 

Anodier mechanism for moisturizing skin is to provide an 
exogenous barrier to water loss (TEWL) so that more water is 
retained in the SC, a "barrier cream". This is the mechanism by 
which petrolatum works, but rather than simply forming a film 
on the skin surface, it has been shown in hairless mice that petro- 
latum penetrates into the intercellular spaces of the SC to provide 
this function (Ghadially et aly 1992). Studies in hairless mice have 
also shown that certain combinations of SC lipids in optimal ra- 
tios can accelerate restoration of barrier function following tape 
stripping or acetone abrogation of the barrier (Mao-Quiang et aly 
1995, 1996). Importantly, certain lipid mixes actually inhibited re- 
storation of barrier function, which has implications for the de- 
sign of topical products. There are relatively few studies on 
human skin of products containing the appropriate mix of SC 
lipids, with varying results (Loden and Barany, 2000; Chamlin 
et aly 2002). Additional well-controlled trials in defined human 
populations are needed to determine if mixtures of SC lipids are 
superior to other formulations. 

In mouse skin, it is argued that topically applied lipids can 
permeate to the granular layer where they become part of LG 
membranes and are then extruded into the intercellular space to 
form the intercellular lamellae (Mao-Quiang et aly 1995). These 
findings, however, need to be confirmed in additional animal 
models and extended to human skin. It has also been shown that 
a mixture of SC lipids applied to lipid-extracted SC sheets can 
restore barrier function (Onken and Moyer, 1963; Imokawa et aly 
1991) and reform intercellular lamellae (Imokawa et al, 1991) and 



238 MADISON 



THE JOURNAL OF INVESTIGATIVE DERMATOLOGY 



that mixtures of SC lipids can form bilayer structures in vitro 
(Kuempel et al, 1998). This suggests that in situ formation of barrier 
lipid membranes by topically applied lipids is another possible 
mechanism of barrier restoration. Ultimately, the goal is to be 
able to tailor topical products to the needs of patients based on 
specific knowledge of their underlying epidermal defects and a 
more complete understanding of how these products work at 
the molecular level. 



OTHER COMPONENTS OF EPIDERMAL BARRIER 
FUNCTION 

Although tight junctions have been sporadically observed in epi- 
dermis for many years, and tight junction proteins are expressed 
in epidermis (Morita et al, 1998), it was not until the recent devel- 
opment of a claudin-1 knockout mouse (Furuse etal^ 2002) that the 
functional role of epidermal tight junctions was examined. The 
neonatal knockout mice showed wrinkled skin with markedly in- 
creased TEWL and died within Id of birth. Expression of loricrin, 
involucrin, and transglutaminase were normal, and lamellar lipid 
structures in the LG extrusion zone and the SC appeared normal. 
Whereas it is not surprising, given the importance of the epidermal 
barrier, that there should he more than one mechanism contribut- 
ing to barrier function, it is interesting that neither system is able 
to compensate for defects in the other. As claudin-1 -deficient SC 
appears different on histology and electron microscopy (more 
compact and without a "basketweave" artifact), it will be important 
to conduct more detailed studies of the lipid composition and 
structure in these mice to determine whether or not lipid altera- 
tions contribute to the diminished barrier function. 



CHALLENGES 

Relatively few mechanistic studies of barrier repair or moisturizer 
effects on human skin have been performed. Short-term acetone 
treatment, a commonly used "barrier disrupter" in many models, 
does not extract significant amounts of barrier lipids from normal 
human skin (Onken and Moyer, 1963; Adams et al, 1993) and the 
mechanism of its effect on barrier function has not been suffi- 
ciently investigated. A variety of solvents, including acetone, may 
give spurious TEWL readings (Morrison, 1992; Adams et al, 1993) 
suggesting that solvent-induced barrier dysfunction may be a less 
desirable model Other models of barrier disruption, both experi- 
mentally induced and natural (due to disease, genetic manipula- 
tion, or environmental conditions) are available, but there 
remains a need for more sophisticated and controlled methods to 
disrupt barrier function. Because TEWL data are affected by a 
number of variables, including varying by up to about 20% based 
on circadian rhythms alone (Le Fur et al, 2001), measurements ob- 
tained after experimental manipulation of the skin require very 
careful interpretation (McCallion and Li Wan Po, 1995; Orth and 
Appa, 2000) with attention to physiologic and clinical versus statis- 
tical significance. In fact, recent studies by Chilcott et al (2002) 
showed no correlation between measured TEWL and skin barrier 
function. Although these rather startling findings need to be du- 
plicated by other investigators and in other models, the authors 
correctly conclude that further work needs to be done on the in- 
terpretation of TEWL. It is best for studies of SC function to 
measure multiple parameters including clinical appearance, 
TEWL, SC water content, permeabiUty to exogenous substances, 
lipid content and composition (and ideally synthesis), and mor- 
phology. Although the use of ruthenium tetroxide postfixation 
in electron microscopy allows visualization of the SC lipid lamel- 
lae and is now widely used to assess changes in lamellar structure 
under a variety of conditions, this technique is far from optimal 
for overall preservation of tissue architecture and is known to re- 
sult in numerous artifacts (Swartzendruber et al, 1995). For this rea- 
son, establishing true differences between normal versus abnormal 



tissue or treated versus untreated tissue using this technique can be 
problematic. For both osmium- and ruthenium-fixed samples, it 
is best to have the microscopist blinded to the status of the tissues 
if possible, multiple samples and sections must be examined, and 
nondramatic findings of difference quantitated wherever possible. 



FUTURE DIRECTIONS 

Studies of epidermal differentiation in keratinocyte culture mod- 
els and in fetal development models will continue to improve our 
understanding of the mechanisms underlying SC barrier forma- 
tion and how they are regulated. Biophysical studies of SC lipids 
will delineate the molecular basis for their barrier properties. So- 
phisticated cell-free in vitro systems will be developed to study la- 
mellar "granule" assembly and the membrane dynamics involved 
in the formation of SC lamellae. Transgenic and knockout mice 
with unexpected barrier defects will be helpful in deHneating 
new regulatory pathways in barrier formation. The development 
of rodent models with defects in specific aspects of barrier lipid 
metabolism will improve our understanding of barrier disrup- 
tion, as we have already seen with p-glucocerebrosidase-deficient 
mice and acid sphingomyelinase-deficient mice. Careful interpre- 
tation and clinical application of all of these findings will greatly 
improve the specificity and efficacy of treatments for human skin 
with abnormal SC structure and function. 



Thanks to my co-investigator Gloria N. Sandoj to Donald C. Swartzendruber for 
providing the electron micrographs, to Elizabeth J. Howard for her technical contribu- 
tions and to Lou Messetie for drawing the ceramide structures. Special thanks to Philip 
W. Wertz for his ongoing support, critical discussions and suggestions related to the field 
of epidermal barrier function. Our work has been supported by The Dermatology 
Foundatiort, NIAMS (NIH), and by a bequest from the Carl J. Herzog Foundation, 
My apologies to all investigators in this area whose publications I did not cite in this 
overview. By necessity, this is only a selection of the many outstanding papers available 
in the literature. Updated from the Dermatology Foundation, Progress in Dermatology, 
December, 2000 



REFERENCES 

Abraham W, Wertz PW, Downing DT: Unoleate-rich acylglucosylceramides of pig 
epidermis: Structure determination by proton magnetic resonance. J LipiV Res 
26:761-765, 1985 

Abraham W, Wertz PW, Landmann L, Downing DT: Stratum comeum lipid lipo- 
somes: Calcium-induced transformation into lamellar sheets. J Intfest Dermatol 
88:212-214, 1987 

Adams K, Harvell JD, Shriner D, Wertz P, Maibach H, Maibach HI, Rehfeld SJ: Ef- 
fect of organic solvents on in vitro human skin water barrier (unction. J Inwst 
Dewtatol 101:609-613, 1993 

Arnold M-L, An ton-La mprccht I, Melz-Rothfuss B, Hartschuh W: Idithosis conge- 
nita type III. Clinical and ultrastructural charaaeristics and distinction within 
the heterogeneous ichthyosis congenita group. Arch Dermatol Res 280:268-278, 
1988 

Attar PS, Wertz P'^ McArthur M, Umakado S. Bickenbach JR, Roop DR: Inhibi- 
tion of retinoid signaling in transgenic mice alters lipid processing and disrupts 
epidermal barrier function. Mol Endocrinol 11:792-800, 1997 

Attenborough D: Life on Earth. Boston: Little, Brown, 1980 

Berenson OS, Burch GE: Studies of diffusion of water through dead human skin: 
The effect of different environmental states and of chemical alterations of the 
epidermis. Am J Trop Med 31:842-853, 1951 

Blank IH; Factors which influence the water content of the stratum comeum. J Invest 
Demxatol 18:433^*40, 1952 

Blank IH; Further observations on factors which influence the water content of the 
stratum comeum. J Jwej( Dermatol 45:249-256, 1953 

BouwstraJA, Gooris GS, Dubbelaar FE, Ponec M: Cholesterol sulfate and calcium 
affect stratum comeum lipid organization over a wide temperature range: J Lj- 
pirf R« 40:2303-2312, 1999 

Bouwstra J A, Dubbelaar FER, Gooris GS, Ponec M: The lipid organization in the 
skin barrier. Acta Demi Venereol 208:23-30, 2000 

Bouwstra J A, Gooris GS, Dubbelaar FE, Ponec M; Phase behavior of stratum cor- 
neum lipid mixtures based on human ceramides; Hie role of natural and syn- 
thetic ceramide 1. J Invest Dermatol 118:606-617, 2002 



VOL. 121, NO. 2 AUGUST 2003 



EPIDERMAL BARRIER FUNCTION 239 



BouwstraJA, Honeywell-Nguyen L, Gooris GS, Ponec M: Structure of the skin bar- 
rier and its modulation by vesicular formulations. Prog Lipid Res 42:1-36, 2003 

Brattsand M, Egelrud T; Purification, molecular cloning, and expression of a human 
stratum comeum trypsin-like serine protease with possible function in desqua- 
mation. J Biol Chem 274:30033-30040, 1999 

Bravcrman N, Lin P, Moebius FF, et al: Mutations in the gene encoding 3 beta-hy- 
droxy steroid-delta 8, delta 7-isomerase cause X-linked dominant Conradi- 
Hunermann syndrome. Nat Genet 22:291-294, 1999 

Breathnach AS, Goodman T, Stolinski C, Gross M: Freeze-fracture replication of 
cells in stratum comeum of human epidermis. J ^rwf 114:65-81, 1973 

Cabral A, Saylin A, dc Winter S, Fischer DF, Pavel S, Backendorg C: SPRR4, a 
novel comified envelope precursor: UV-dependent epidermal expression and 
selective incorporation into fragile envelopes. J Q// Sci 114:3837-3843, 2001 

Chamlin SL, Kao J, Frieden IJ, et al: Ceramide-dominant barrier repair lipids allevi- 
ate childhood atopic dermatitis. Changes in barrier function provide a sensitive 
indicator of disease activity. J /Im Acad Demiatol 47:198-208, 2002 

Chavanas S, Bodemar C, Rochat A, et al: Mutations in S PINKS, encoding a serine 
protease inhibitor, cause Netherton syndrome. Nat Genet 25:141-142, 2000 

Chilcott RP, Dalton CH, Emmanuel AJ, Allen CE, Bradley ST: Transepi dermal 
water loss does not correlate with skin barrier function in vitro. J Invest 
Dermatol 118:871-875,2002 

De Lauren zi V, Rogers OR, Hamrock DJ, et al: Sjogren-Larsson syndrome is caused 
by mutations in the fatty aldehyde dehydrogenase gene. Nat Genet 12:52-57, 
1996 

Denda M, Sato J,TsuchiyaT, EUas PM, Feingold KR: Low humidity stimulates epi- 
dermal DNA synthesis and amplifies the hyperprolifcrative response to barrier 
disruption: Implication for seasonal exacerbations of inflammatory dermatoses. 
J Invest Dermatol 111:87^78, 1998 

Djian P, Easley K, Green H: Targeted ablation of the murine involucrin gene. J Cell 
Bio/ 151:381-387, 2000 

DoeringT, Holleran WM, Potratz A,Vielhaber G, Elias PM, Suzuki K, Sandhoff K: 
Sphingolipid aaivator proteins are required for epidermal permeability barrier 
formarion. JBib/ Chem 274:11938-11045, 1999a 

Doering T, Proia RL, Sandhoff K: Accumulation of protein-bound epidermal glu- 
cosylceramides in beta-^lucocerebrosidase deficient type 2 Gaucher mice. 
FEBS Lett 447:167-170. 1999b 

DoeringT, Brade H, Sandhoff K: Sphingolipid metabolism during epidermal barrier 
development in mice. J Lipid Res 43:1727-1733, 2002 

Downing DT; Lipid and protein structures in the permeability barrier of mam- 
malian epidermis. J Lipid Res 33:301-313, 1992 

Elias PM, Friend DS: The permeability barrier in mammalian epidermis. J Cell Biol 
65:180-191, 1975 

Elias PM, McNutt NS, Friend DS: Membrane alterations during comification of 
mammalian squamous epithelia: A freeze-fracture, tracer, and thin-section 
smdy. Anat Rec 189:577-594, 1977 

Elias PM, Cullander C, Mauro X Rassner U, Komuves L, Brown BE. Menon GK: 
The secretory granular cell: The outermost granular cell as a specialized secre- 
tory cell J Invest Dermatol Symp Proc 3:87-100, 1998 

Elias PM, Holleran WM, Calhoun CJ, Quiec D, Brown BE, Behnc M. Feingold 
KR: Permeability barrier homeostasis. The role of lipid processing. In: Loden 
M, Maibach HI (eds). Dry Skin and Moisturizers: Chemistry and Function. Derma- 
tology: Clinical and Basic Science Series. New York; CRC Press, 2000; p 59- 
70 

Elias PM, Schmuth M, Uchida Y, et al: Basis for the permeability barrier abnormality 

in lamellar ichthyosis. Exp Dermatol 11:248-256, 2002 
Fartasch M, Williams ML, Elias PM: Altered lamellar body secretion and stratum 

comeum membrane structure in Netherton syndrome: Differentiation from 

other infantile erythrodermas and pathogenic implications. Arch Dermatol 

135:823-832, 1999 

Feingold KR, Elias PM: The environmental interface. Regulation of permeability 
barrier homeostasis. In: Loden M, Maibach HI (eds). Dry Skin and Moisturizers: 
Chemistry and Function. Dermatology: Clinical and Basic Science Series. New 
York: CRC Press, 2000; p 45-58 

Fluhr JW, Gloor M, Lehmann L, Lazzerini S, Distante F, Berardesca E: Glycerol ac- 
celerates recovery of barrier function in vivo. Acta Demi Venereol 79:418—421, 
1999 

ForsHnd B: A domain mosaic model of the skin barrier. Acta Derm Venereol 74:1-6, 
1994 

Freinkd RK, Traczyk TN: Lipid composition and acid hydrolase content of lamellar 
granules of fetal rat epidermis. J Invest Dermatol 85:295-298, 1985 

Furuse M, Hata M, Furuse K, et ah Claudin-based tight junctions are crucial for the 
mammalian permeability barrier A lesson from daudin-1 -deficient mice. 
J Cell Biol 156:1099-1111, 2002 

Ghadially R, Brown BE, Sequeira-Martin SM, et ah The aged epidermal 
permeability barrier: Structural, functional and lipid biochemical 
abnormalities in humans and a senescent murine model J C//n Invest 95:2281- 
2290, 1995 

Ghadially R, Halkier-Sorensen L, EHas PM: Effects of petrolatum on stratum cor- 
neum structure and function. _//lttr Acad Dermatol 26:387-396, 1992 

Ghadially R, Brown BE, Hanley K, et al: Decreased epidermal lipid synthesis ac- 
counts for altered barrier function in aged mice. J Invest Dermatol 106:1064- 
1069, 19%a 



Ghadially R, ReedJT, Elias PM: Stramm comeum structure and function correlates 
with phenotype in psoriasis. J Invest Dermatol 107:558-564, 1996b 

Grange DK, Kratz LE, Braverman NE, Kelley Rl: CHILD syndrome caused by de- 
ficiency of 3 beta-hydroxysteroid-delta 8, delta 7-isomerase. Am J Med Genet 
90:328-335, 2000 

Grayson S, Johnson-Winegar AG, Wintroub BU, IsseroffRR, Epstein EH, Elias PM: 
Lamellar body-enriched fractions from neonatal mice, preparative techniques 
and partial characterization. J Invest Dermatol 85:289-294, 1985 

Hamanaka S. Hara M, Nishio H, Otsuka F, Suzuki A, Uchida Y: Human epidermal 
glucosylccramides are major precurson of stratum comeum ceramides. J Invest 
DemjflM/ 119:416-423, 2002 

Hanley K, Komuves LG, Bass NM, et al: Fetal epidermal differentiation and barrier 
development in vivo is accelerated by nuclear hormone receptor activators. J 
Invest Dermatol 113:788-795, 1999 

Hansson L, Stromqvist M, Backman A,Wallbrandt P, Carlstein A, Egelrud T: Clon- 
ing, expression, and characterization of stratum corneum chymotrypric en- 
zyme, a skin-specific human serine proteinase. J Biol Chem 269:19420-19426, 
1994 

Hansson L, Backman A, Ny A, e( al: Epidermal overexpression of stramm corneum 
chymotrypric enzyme in mice: A model for chronic itchy dermatitis. J Invest 
Dermatol 118:444-449, 2002 

Hara J, Higuchi K, Okamoto R, Kawashima M, Imokawa G: High-expression of 
sphingomyelin deacylase is an important determinant of ceramide deficiency 
leading to barrier disruption in atopic dermatitis. J Invest Dermatol 115:406-413, 
2000 

Harding CR, Bartolone J, Rawlings AV: Effects of namral moisturizing factor and 
lactic acid isomers on skin function. In: Loden M, Maibach HI (eds). Dry Skin 
and Moisturizers: Chemistry and Function. Dermatology: Clinical and Basic 
Science Series. New York: CRC Press, 2000; p 229-241 

Hart TC, Hart PS, Bowden DW, et al: Mutations of the cathepsin C gene are respon- 
sible for Papillon-Lefevre syndrome. J Med Genet 36:881-887, 1999 

Hedberg CL, Wertz PW, Downing DT; The time course of lipid biosynthesis in pig 
epidermis. J /«t;e5( Dermatol 91:169-174, 1988 

Hirschberg K, Miller CM, EllenbergJ, Presley JF, Siggia ED, Phair RD, Lippincott- 
Schwartz J: Kinetic analysis of secretory membrane traffic and characterization 
of Golgi to plasma membrane transport intermediates in living cells. J Cell Biol 
143:1485-1503. 1998 

Holleran WM, Ginns El, Menon GK, etah Consequences of beta-glucocerebrosidase 
deficiency in epidermis. Ultrastructure and permeability barrier alterations in 
Gaucher disease. J Clin Invest 93:1756-1764, 1994 

Horikoshi T, Arany I, Rajaraman S, et al: Isoforms of cathepsin D and human epi- 
dermal differentiation. Biochimie 80:605-612, 1998 

Imokawa G, Kuno H, Kawai M: Stratum comeum lipids serve as a bound-water 
modulator. J Invest Dermatol 96:845-851, 1991 

Jain A, Rutter N, Cartlidge PH: Influence of antenatal steroids and sex on mamra- 
tion of the epidermal barrier in the preterm infant. Arch Dis Child Fetal Neotm- 
to/Ed83:H12-F116, 2000 

Jansen GA, Ofman R, Ferdinandusse S, et ah Refsum disease is caused by mutations 
in the phytanoyl-CoA hydroxylase gene. Nat Genet 17:190-193, 1997 

Jamik M, de Viragh PA, Scharer E, Bundman D, Simon MN. Roop DR^ Steven 
AC: Quasi-normal comified cell envelopes in loricrin knockout mice 
imply the existence of a loricrin backup system. J Invest Dermatol 118:102-109, 
2002' 

Jensen J-M, Schutze S, Fori M, Kronke M, Proksch E: Roles for tumor necrosis fac- 
tor receptor p55 and sphingomyelinase in repairing the cutaneous permeability 
barrier. J Clin Invest 104:1761-1770, 1999 

Jetten AM, George MA, Nervi C, Boone LR, Rearick JI: Increased cholesterol sul- 
fate and cholesterol sulfotransferase activity in relation to the multi-step pro- 
cess of differentiation in human epidermal kerarinocytes. J Invest Dermatol 
92:203-209, 1992 

Jobard F, Lefevre C, Karaduman A, et ah Lipoxygenase-3 (ALOXE3) and 12 (R)-li- 
poxygenase (ALOX12B) are mutated in non-bullous congenital ichthyosiform 
erythroderma (NCIE) linked to chromosome 17pl3.1. Hum Mol Genet 11:107- 
113. 2002 

Kajava AV: Alpha-helical solenoid model for the human involucrin. FEBS Lett 
473:127-131, 2000 

Kali YN, Nonato LB. Lund CH, Guy RH: Development of skin barrier function in 
premature infants. J Invest Dermatol 111:320-326. 1998 

Kalinin AE, Kajava AV, Steincrt PM: Epithelial barrier function: Assembly and 
structural features of the comified cell envelope: Bioessays 24:789-800, 2002 

Koch PJ, deViragh PA, Scharer E, et ah Lessons from loricrin-deficient mice: Com- 
pensatory mechanisms maintaining skin barrier function in the absence of a 
major comified envelope protein. J Q// Biol 151:38^00, 2000 

Komatsu N, Takata M, Otsuki N, Ohka R, Amano O, Takehara K, Saijoh K: Ele- 
vated stratum corneum hydrolyric aaivity in Netherton syndrome suggests 
an inhibitory regulation of desquamation by SPINK5 -derived peptides. J In- 
vest Dermatol 118:436-443. 2002 

Konig A, Hap pie R, Bomholdt D, Engel H, Grzeschik K-H: Mutations in the 
NSDHL gene, encoding a 3beta-hydroxy steroid dehydrogenase, cause CHILD 
syndrome. Am J Med Genet 90:339-346, 2000 

Kuempel D, Swartzendruber DC, Squier CA, Wertz PW; In vitro reconstitution of 
stratum comeum lipid lamellae. Biochim Biophys Acta 1372:135-140, 1998 



240 MADISON 

Landmann L: Epidermal permeability barrier: Transformation of lamellar granule 
disks into intercellular sheets by a membrane fusion process. J Invest Dermatol 
87:202-209, 1986 

Landmann L:The epidermal permeability barrier. Amt Embryol 178:1-13, 1988 
Landmann L, Wertz P"W, Downing DT: Acylglucosylceramide causes flattening and 

stacking of liposomes: An analogy for assembly of the epidermal permeability 

barrier. Biochim Biophys Acta 778:412-418, 1984 
Lavrijsen AP, BouwstraJA, Goons GS, Weerheim A, Bodde HE, Ponec M: Reduced 

skin barrier function parallels abnormal stratum comeum lipid o^anizarion in 

patients with lamellar ichthyosis. J Invest Dermatol 105:619-624, 1995 
Lazo ND, Downing DT: A mixture of alpha-helical and 310-helical conformations 

for involucrin in the human epidermal comeocyte envelope provides a scaffold 

for the attachment of both lipids and proteins. J Biol Chem 274:37340-37344, 

1999 

Le Fur I, Reinberg A, Lopez S, Morizot F, Mechkouri M.Tschachler E: Analysis of 
circadian and ultradian rhythms of skin surface properties efface and forearm 
of healthy vfomtn. J Invest Demtatol U7:7l&-724, 2001 

Lee SH, Choi EH, Feingold KR, Jiang S, Ahn SK: Iontophoresis itself on hairless 
mouse skin induces the loss of the epidermal calcium gradient without skin 
barrier impairment. J Invest Dermatol 111:39-43, 1998 

Lefevre C, Jobard F, Caux F, et al: Mutations in CGl-58, the gene encoding a new 
protein of the esteras^ipase/thioestserase subfamily, in Chanarin-Dorfman 
syndrome. Am J Hum Genet 69:1002-1012, 2001 

Loden M, Barany E: Skin-identical lipids versus petrolatum in the treatment of tape- 
stripped and detergent-perturbed human skin. Acta Derm Venereol 80:412-415, 
2000 

Long SA, Wertz PW, Strauss JS, Downing DT: Human stratum comeum polar lipids 

and desquamation. Arch Dermatol Res 277:284-287, 1985 
Macheleidt O, Kaiser HW, SandhofF K: Deficiency of epidermal protein-bound 

omega-hydroxyceramides in atopic dermatitis. J Invest Dermatol 119:166-173, 

2002 

Madison KC, Howard £J; Ceramides are transported through the Golgi apparatus in 
keratinocytes in vitro. J /ni/ejf Dermatol 106:1030-1035. 1996 

Madison KC, Swartzendruber DC, Wertz PW, Downing DT: Presence of intact in- 
tercellular lipid lamellae in the upper layers of the stratum comeum. J Invest 
Dermatol 88:714-718, 1987 

Madison KC, Swartzendruber DC, Wertz PW, Downing DT: Lamellar granule ex- 
trusion and stratum comeum intercellular lamellae in murine keratinocyte cul- 
tures. J Invest Dermatol 90:110-116, 1988 

Madison KC, Swartzendruber DC, Wertz PW, Downing DT: Murine keratinocyte 
cultures grown at the air/medium interface synthesize stratum comeum 
lipids and "recycle" linoleate during differentiation. J Invest Dermatol 93:10-17, 
1989 

Madison KC, Swartzendmber DC, Wertz PW, Downing DT: Sphingolipid metabo- 
lism in organotypic mouse keratinocyte cultures. J Invest Dermatol 96:657-664, 
1990 

Madison KC, Sando GN, Howard EJ, True CA, Gilbert D, Swartzendruber DC, 
Wertz PW: Lamellar granule biogenesis. A role for ceramide glucosyltransfer- 
ase, lysosomal enzyme transport, and the Golgi. J Invest Demtatol Symp Proc 
3:8(>-«6, 1998 

Mancini AJ, Sookdeo-Drost A, Madison KC, SmoUer BR, Lane AT: Semipermeable 
dressings improve epidermal barrier funaion in premature infants. Pediatr Res 
36:306-314, 1994 

Mao-Quiang M, Brown BE,Wu-Pong S, Feingold KR, Elias PM: Exogenous non- 
physiologic vs physiologic lipids. Divergent mechanisms for correction of per- 
meability barrier dysfunction. Arch Dermatol 131:809-816, 1995 

Mao-Quiang M, Feingold KR, Thornfeldt CR, Elias PM: Oprimization of physio- 
logical lipid mixtures for barrier repair. J Demtatol 106:1096-1101, 1996 

Marekov LN, Steinert PM: Ceramides are bound to structural proteins of the human 
foreskin epidermal comified cell envelope: J Biol Chem 273:17763-17760, 1998 

Marshall D, Hardman MJ, Nield KM, Byrne C: Differentially expressed late consri- 
tuents of the epidermal comified envelope: Proc Natl Acad Sci USA 98:B031- 
13036, 2001 

Matoltsy AG, Downes AM, Sweeney TM; Studies of the epidermal water barrier. 
Part II. Investigation of the chemical nature of the water barrier J Invest Demta- 
tol 50:19-26,1968 

McCallion R, Li Wan Po A: In vivo evaluation of the effects of moisturisers on 
transepidermal water loss using faaorial designs, Int J Phamtaceutics 113:247- 
255, 1995 

Melton JL, Wertz PW, Swartzendruber DC, Downing DT: Effects of essential fatty 
acid deficiency on O-acylsphingolipids and transepidermal water loss in young 
pigs. Biochim Biophys Acta 921:191-197, 1987 

Menon GK, Grayson S, Elias PM: Cytochemical and biochemical localization of li- 
pase and sphingomyelinase activity in mammalian epidermis. J /ni/cif Demtatol 
86:591-597, 1986 

Menon GK, Feingold KR, Elias PM: Lamellar body secretory response to barrier 
disruption. J Invest Demtatol 98:279-289, 1992 

Milner ME, O'Guin WM, Holbrook KA, Dale BA: Abnormal lamellar granules in 
harlequin ichthyosis. J /nmr Demtatol 99:824-829, 1992 

Mironov A A, Luini A, Buccione R: Constitutive transport between the trans-Golgi 
network and the plasma membrane according to the maturation model. A hy- 
pothesis. FEBS Lett 440:99-102, 1998 



THE JOURNAL OF INVESTIGATIVE DERMATOLOGY 

Morita K, Itoh M, Saitou M, et al: Subcellular distribution of tight junaion-asso- 
ciated proteins, occludin, ZO-1 and ZO-2, in rodent skin and its developmen- 
tal changes J Invest Dermatol 110:862-866, 1998 

Morrison BM: ServoMed evapori meter: Precautions when evaluating the effect of 
skin care products on barrier function. J Sac Cosmet Chem 43:161-167, 1992 

Motta S, Monti M, Sesana S, Caputo R, Carelli S, Ghidoni R: Ceramide composi- 
tion of the psoriatic scale. Biochim Biophys Acta 1182:145-151, 1993 

Motta S, Monti M, Sesana S, Mellesi L, Ghidoni R, Caputo R: Abnormality of 
water barrier function in psoriasis: Role of ceramide fractions. AkH Dermatol 
130:452-456, 1994 

Nemes Z, Demeny M, Marekow LN, Fesus L, Steinert P: Cholesterol 3-sulfate inter- 
feres with comified envelope assembly by diverting transglutaminase 1 activity 
from the formation of cross-links and esters to the hydrolysis of glutamine: J 
Biol Chem 275:2636-2646, 2000 

Nemes Z, Marekov LN, Fesus L, Steinert PM: A novel function for transglutaminase 
1. Attachment of long-chain omega-hydroxyceramides to involucrin by ester 
bond formation. Proc Natl Acad Sci USA 96:8402-8407, 1999 

Norlen L: Skin barrier formation: The membrane folding model. J Invest Dermatol 
117:823-829, 2001a 

Norlen L: Skin barrier structure and function: The single gel phase model J Invest 

Dermatol 117:83(>-«36, 2001b 
Onken HD, Moyer CA: The water barrier in human epidermis. Anh Dermatol 

87:584-590, 1963 

Orth DS, Appa Y: Glycerine: A natural ingredient for moisturizing skin. In: Loden 
M, Maibach HI (eds). Dry Skin and Moisturizers: Chemistry and Function. Derma- 
tology: Clinical and Basic Science Series. New York: CRC Press, 2000; p 213- 
228 

Pilgram GS, Vissers DC, van der Meulen H, Pavel S, Lavrijsen SP, Bouwstra JA: 
Aberrant lipid organization in stratum comeum of patients with atopic derma- 
titis and lamellar ichthyosis. J Invest Demtatol 117:710-717, 2001 

Ponec M, Weerheim A, Kempcnaar J, Mulder A, Gooris GS, Bouwstra J, Mommaas 
AM: The formation of competent barrier lipids in reconstmcted human epi- 
dermis requires the presence of viumin C J Invest Dermatol 109:348-355, 1997 

Ponec M, Weerheim A, Lankhorst P, Wertz P: New acylceramide in native and re- 
constructed epidermis. J /m/«f Demtatol 120581-588, 2003 

Rigg PC, Barry BW: Shed snake skin and hairless mouse skin as model membranes 
for human skin during permeation studies. J Invest Dermatol 94:235-240, 1990 

Robson KJ, Stewart ME, Michelsen S, Lazo ND, Downing DT: 6-Hydroxy-4- 
sphingenine in human epidermal ceramides. J Lipid R£S 35:2060-2068, 1994 

Sando GN, Howard EJ, Madison KC: Induction of ceramide glucosyltransferase in 
cultured keratinocytes: Correlation with culture differentiation. J Biol Chem 
271:22044-22051, 1996 

Sato J, Denda M, NakanishiJ, Nomura J, KlyamaJ: Cholesterol sulfate inhibits pro- 
teases that are involved in desquamation of stratum comeam. J Invest Dermatol 
111:189-193, 1998 

Scheuplein R, Ross L: Effects of surfactants and solvents on the permeability of epi- 
dermis, J Sac Cosmet Chem 21:853-873, 1970 

Schmuth M, Man M-Q Weber F, Gao W, Feinfold KR, Fritsch P, Elias PM, Holleran 
WM: Permeability barrier disorder in Niemann-Pick disease: Sphingomyelin- 
ceramide processing required for normal permeability barrier homeostasis. J 
Invest Dermatol 115:459-466, 2000 

Schmuth M,Yosipovitch G, Williams ML, etah Pathogenesis of the permeability bar- 
rier abnormality in epidermolyric hyperkeratosis. J Invest Dermatol 117:837-847, 
2001 

Schmuth M, Schoonjans K.Yu Q-C et al: Role of peroxisome proliferator-activated 
receptor alpha in epidermal development in utero. J Invest Dermatol 119:1298- 
1303, 2002 

Schreiner V, Gooris GS, Pfeiffer S, et al: Barrier characteristics of different human 
skin types investigated with X-ray diffraction, lipid analysis, and electron mi- 
croscopy imaging. J Invest Dermatol 114:654-660, 2000 

Segre J A, Bauer C, Fuchs E: Klf4 is a transcription factor required for establishing 
the barrier function of skin. Nature Genet 22:356-360, 1999 

Shapiro LJ, Weiss R, Webster D, France JT: X-linked ichthyosis due to steroid sulfa- 
tase deficiency. Lancet i:70-72, 1978 

Sidransky E, Sherer DM, Ginns EI: Gaucher disease in the neonate. A distinct Gau- 
cher phenotype is analogous to a mouse model created by Urgeted disruption 
of the beta-glucocerebrosidase gent Pediatr Res 32:494-498, 1992 

Siegfried EC: Neonatal skin and skin care. Demtatol Clin 16:437-446, 1998 

Silverman RA, Lender J, Elmets CA: Effects of occlusive and semiocclusive dressings 
on the return of barrier function to transepidermal water loss in sundardized 
human wounds. J ylni Acad Demtatol 20:755-760, 1989 

Simon M, Jonca N, Guerrin M, et ah Refined charaaerization of comeodesmosin 
proteolysis during terminal differentiation of human epidermis and its rela- 
tionship to desquamation. J Biol Chem 276:20292-20299, 2001 

Sondell B, Thornell L-E, Egelrud T: Evidence that stratum comeum chymotryptic 
enzyme is transported to the stratum comeum extracellular space via lamellar 
bodies. J Invest Demtatol 104:819-823, 1995 

Stachowitz S, Alessandrini F, Abeck D, Ring J, Behrendt H: Permeability barrier 
disruption increases the level of serine palmitoyltransferase in human epider- 
mis. J Invest Demtatol 119:104&-1052, 2002 

Steinert PM: The complexity and redundancy of epithelial barrier function. J Cell 
Biol 151;F5-F7, 2000 



VOL 121, NO. 2 AUGUST 2003 



EPIDERM AL BARRIER FUNCTION 241 



Stewart ME, Downing DT: A new 6-hydroxy-4-sphingenine-containing ceramide 
in human skin. J Lipid Res 40:1434-1439, 1999 

Stewart ME, Downing DT: The omega-hydroxyceramides of pig epidermis are at- 
tached to comeocytes solely through omega-hydroxyl groups. J Lipid Res 
42:1105-1110, 2001 

Swartzendruber DQWertz PW, Madison KC, Downing DT: Evidence that the cor- 
neocyte has a chemically bound lipid envelope. J Invest Dermatol 88:709-713, 
1987 

Swartzendruber DC, Wertz PW, Kitko DJ, Madison KC, Downing DT; Molecular 
models of the intercellular lipid lamellae in mammalian stratum comeum, J 
Invest Dermatol 92:251-257, 1989 

Swartzendruber DC, Burnett IH, Wertz PW Madison KC, Squier CA: Osmium 
tetroxide and ruthenium tetroxide are complementary reagents for the pre- 
paration of epidermal samples for transmission electron microscopy. J Invest 
Demiatol 104:417-420, 1995 

Sweeny TM, Downing DTrThe role of lipids in the epidermal barrier to water diffu- 
sion. J Invest Dermatol 55:135-140, 1970 

Sybert VP, Dale BA, Holbrook KA: Ichthyosis vulgaris: Identification of a defect in 
filaggrin synthesis correlated with an absence of keratohyalin granules. J /fiver( 
Dermatol 84:191-194, 1985 

Van de Kerkhof PCM, De Mare S, Arnold WP, Van Brt PEJ; Epidermal regeneration 
and occlusion. Acta Derm Venereol 75:6-8, 1995 

Virolainen E, Niemi KM, Ganemo A, Kere J, Vahlquist A, Saarialho-Kere U: Ultra- 
structural features resembling those of harlequin ichthyosis in patients with se- 
vere congenital ichthyosiform erythroderma, Br J Dermatol 145:480-483, 2001 

Walley AJ, Chavanas S, Monatt MF, et d: Gene polymorphism in Netherton and 
common atopic disease. Nat Genet 29:175-178, 2001 

Watanabe R, Wu K, Paul P, Marks DL, Kobayashi T, Pittelkow MR, Pagano RE: 
Up-regulation of glucosylccramide synthase expression and activity during 
human keratinocy te differentiation. J Biol Chem 273:9651-9655, 1998 

Welzel J, Wilhelm K-P, Wolff HH: Occlusion does not influence the repair of the 
permeability barrier in human skin. In: Eisner P, Maibach HI (eds). Irritant Der- 
matitis. New Clinical and Experimental Aspects. Current Problems in Dermatology 
Vol 23. Basel: Karger, 1995; p 180-186 



Welzel J, Wilhelm K-P.WolfTHH: Skin permeability barrier and occlusion: No delay 
of repair in irritated human skin. Contact Dermatitis 35:163-168, 1996 

Wertz PW: The nature of the epidermal barrier: Biochemical aspects. Adv Drug Del 
Rev 18:283-294, 1996 

Wertz PW, Downing DT: Glycolipids in mammalian epidermis. Structure and func- 
tion in the water barrier. Science 217:1261-1262, 1982 

Wertz PW, Downing DT: Ceramides of pig epidermis: Structure determination. 
J Lipid Res 24:759-765, 1983 

Wertz PW, Downing DT: Covalently bound omega-hydroxylacylsphingosine in the 
stratum comeum. Biochim Biophys Acta 917:108-111, 1987 

Wertz PW, Downing DT: Metabolism of linoleic add in porcine epidermis. J Lipirf 
R« 31:1839-1844,1990 

Wertz PW, Swartzendruber DC, Abraham W, Madison KC, Downing DT: Essential 
fatty acids and epidermal integrity. AkH Dermatol 123:1381-1384, 1987 

Wertz PW, Swartzendruber DC, Kitko DJ, Madison KC, Downing DT: The role of 
the comeocyte lipid envelopes in cohesion of the stratum comeum, ] Invest 
Dermatol 93:169-172, 1989 

Williams ML, Elias PM: Ichthyosis; Where we have been; Disorders of comification: 
Where we are going. Curr Probl Dermatol 12:171-176, 2000 

Williams ML, Hanley K, EHas PM, Feingold KR: Ontogeny of the epidermal per- 
meability barrier. J Invest Dermatol Symp Proc 3:75-79, 1998 

Winsor T, Burch GE: Differential roles of layers of human epigastric skin on diffu- 
sion rate of water. Arch Intern Merf 74:428-436, 1944 

Yardley HJ, Summerly R: Lipid composition and metabolism in normal and dis- 
eased epidermis. Pharmacol Ther 13:357-383, 1981 

Zettersten E, Man M-Q, Sato J, etal: Recessive X-Iinked ichthyosis. Role of choles- 
terol-sulfate accumulation in the barrier abnormality. J Invest Dermatol 111:784- 
790, 1998